- Email: [email protected]
Differences in genotoxicity of H2O2 and tetrachlorohydroquinone in human fibroblasts
Mutation Research 513 (2002) 159–167
Differences in genotoxicity of H2 O2 and tetrachlorohydroquinone in human fibroblasts Martin Purschke, Heike Jacobi, Irene Witte∗ Carl von Ossietzky Universität Oldenburg, FB Biologie and ICBM, Postfach 2503, D-26111 Oldenburg, Germany Received 19 June 2001; received in revised form 30 August 2001; accepted 7 September 2001
Abstract During autoxidation of the pentachlorophenol (PCP) metabolite tetrachlorohydroquinone (TCHQ) the semiquinone is formed as well as reactive oxygen species (ROS). It was examined if • OH or the semiquinone are the cause of TCHQ-induced genotoxicity by direct comparison of TCHQ- and H2 O2 -induced DNA damage in human cells. All endpoints tested (DNA damage, DNA repair, and mutagenicity) revealed a greater genotoxic potential for TCHQ than for H2 O2 . In the comet assay, TCHQ induced DNA damage at lower concentrations than H2 O2 . The damaging rate by TCHQ (tail moment (tm)/concentration) was 10-fold greater than by H2 O2 . DNA repair was lower for TCHQ than for H2 O2 treatment. This was shown by measuring DNA repair in the unscheduled DNA synthesis (UDS) assay and the persistence of the DNA damage in the comet assay. In contrast to H2 O2 , TCHQ in non-toxic concentrations was mutagenic in the hypoxanthine-guanine phosphoribosyltransferase (HPRT) locus of V79 cells. Finally, there were also differences observed in cytotoxicity (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay) of TCHQ and H2 O2 . Whereas the TCHQ cytotoxicity was enhanced during a 21 h recovery phase, the H2 O2 cytotoxicity did not change. The results demonstrated that the pronounced genotoxic properties of TCHQ in human cells were not caused by • OH radicals but more likely by the tetrachlorosemiquinone (TCSQ) radical. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Hydrogen peroxide; Tetrachlorohydroquinone; Mutagenicity; Genotoxicity; DNA repair; Cytotoxicity
1. Introduction Tetrachlorohydroquinone (TCHQ) has been identified as the main DNA damaging metabolite of Abbreviations: cfa, colony forming ability; DMSO, dimethyl sulfoxide; EMS, ethylmethane sulfonate; HPRT, hypoxanthine-guanine phosphoribosyltransferase; MEM, minimal essential medium; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide; PBS, phosphate-buffered saline; PCP, pentachlorophenol; ROS, reactive oxygen species; TCBQ, tetrachlorobenzoquinone; TCHQ, tetrachlorohydroquinone; TCSQ, tetrachlorosemiquinone radical; tm, tail moment; UDS, unscheduled DNA synthesis ∗ Corresponding author. Tel.: +49-441-798-3785; fax: +49-441-789-4903. E-mail address: [email protected] (I. Witte).
the widely used and ubiquitously found wood preservative pentachlorophenol (PCP). TCHQ autoxidizes in two one-electron steps to its tetrachlorosemiquinone (TCSQ) radical and further to its quinone [1,2]. The resulting free electrons reduce oxygen to reactive oxygen species (ROS). Genotoxic events of TCHQ may either be due to ROS or TCSQ. While there is little knowledge on TCSQ-induced DNA damage, ROSinduced DNA damage as a consequence of TCHQ autoxidation has been observed repeatedly. Of the ROS, especially the • OH radical is suspected to be the main damaging species in TCHQ genotoxicity. The formation of the typical • OH radical adduct 8hydroxy-2-deoxyguanosine was detected in vitro [3], in cell culture [4,5] and in mice [6] after incubation
1383-5718/02/$ – see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S 1 3 8 3 - 5 7 1 8 ( 0 1 ) 0 0 3 1 1 - 4
160
M. Purschke et al. / Mutation Research 513 (2002) 159–167
or administration of TCHQ. The complete inhibition of TCHQ-induced strand breaks in PM2 DNA by catalase and the specific scavenger dimethyl sulfoxide (DMSO) [1] pointed to • OH radicals as the only cause of DNA strand breakage in superhelical DNA. Nevertheless, there exists some doubts on the dominant role of • OH radicals in TCHQ-induced genotoxicity. In human fibroblasts, the TCHQ-induced DNA damage measured by the alkaline elution [1] or by the comet assay [7] was not completely inhibited by DMSO, even if high concentrations of DMSO were used. In contrast to TCHQ, • OH radicals generated by H2 O2 induced strand breaks which were totally prevented by DMSO in fibroblasts [8]. Further, for TCHQ or its autoxidation products, the covalent binding to DNA was previously shown [9–12]. The genotoxic consequences of those DNA adducts are unknown so far. From these findings, it should be expected that besides ROS also TCHQ itself or its autoxidation product TCSQ are candidates for inducing genotoxic effects in human cells. The aim of this study was to elucidate in mammalian cells the role of DNA damaging and mutagenic species others than ROS in TCHQ genotoxicity. Therefore, a comparison was made of H2 O2 as an exclusively hydroxyl radical generating molecule and TCHQ which additionally forms the reactive semiquinone and quinone molecule. It is expected that DNA damages and DNA repair induced by TCSQ are different from that of • OH radicals. Especially, DNA repair of TCHQ–DNA adducts should follow other pathways than DNA repair of single strand breaks induced by hydroxyl radicals. DNA damage was measured in human fibroblasts by the comet assay. DNA repair was examined with two methods: the persistence of DNA damage with the comet assay, and the 3 [H]-thymidine incorporation into damaged DNA by the method of the unscheduled DNA synthesis (UDS). The consequences of DNA damage and DNA repair were determined by measuring mutagenicity of H2 O2 and TCHQ in the hypoxanthine-guanine phosphoribosyltransferase (HPRT) assay. In addition, the role of DNA damage in cytotoxicity of H2 O2 and TCHQ was examined by comparing the acute toxicity directly after treatment with the chemicals and after a recovery phase of 21 h. While DNA damage rarely should effect cytotoxicity directly after chemical treatment, a reduced cell
viability is expected hours later if unrepaired DNA damage inhibits the protein biosynthesis. 2. Materials and methods 2.1. Cell cultures Human fibroblasts from apparently normal persons, cell lines GM 5757, GM 5856, GM 5659 were purchased from the Human Genetic Mutant Cell Repository (Camden, NJ). Monolayer cultures (passage 7–15) were grown in Eagle’s minimal essential medium (MEM) supplemented with 12% fetal calf serum, vitamins, essential and non-essential amino acids, with 100 U/ml of both penicillin and streptomycin. The cells were grown at 37 ◦ C in an atmosphere of 5% CO2 and 95% air with more than 95% humidity. 2.2. Chemical treatment of cells Tetrachlorohydroquinone (1 mM; from Sigma, FRG) was freshly dissolved in 100 mM NaPO4 buffer, pH 7.4 and sonicated by ultrasound for 4 min. The TCHQ solution was diluted by serum-free medium to the desired concentrations and immediately after washing the cells were treated with the TCHQ solutions for 1 h at 37 ◦ C. H2 O2 was diluted from a 37% aqueous solution (Acros Organics, NJ) to 5 mM in 100 mM NaPO4 buffer, pH 7.4 immediately before cell treatment. Further dilution occurred in serum-free medium. H2 O2 treatment like TCHQ treatment was performed for 1 h at 37 ◦ C. Thereafter, test solutions were removed and the cells were washed repeatedly. 2.3. Determination of cell viability (MTT assay) Cell viability was measured by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay which was performed according to [13]. The test is based on the reduction of the soluble yellow MTT tetrazolium salt to a blue soluble formazan produced by mitochondrial succinate dehydrogenase. In all, 8000 cells were seeded in each well of a 96-well tissue culture microtiter plate. After cells had achieved confluency, the medium was removed and the cells were treated with the test solutions as described before. At least four wells were used for
M. Purschke et al. / Mutation Research 513 (2002) 159–167
every concentration tested. After 1 h of incubation with TCHQ or H2 O2 , the test solutions were removed and the cells were washed twice. MTT containing medium (100 l), which was freshly prepared by a 1:5 dilution of an MTT stock solution (5 mg/ml phosphate-buffered saline (PBS) buffer) was added to each well. The plate was incubated for an additional 3 h. An amount of 100 l of lysis solution (20 g/l SDS, 50%, v/v DMF, pH adjusted to 4.7 by adding 2.5% (v/v) of a 80% acetic acid and 2.5% 1N HCl) were added to each well and the plate was shaken overnight in the dark to extract and solubilize the formazan. Formation of formazan was measured with a Biorad microplate reader using a 570 nm test wavelength and a 655 nm reference wavelength. Cell viability was calculated as the percent ratio of the absorbance of the samples to the referring control. 2.4. Determination of DNA strand breakage (comet assay) The comet assay was performed according to Singh et al. [14]. In brief, 30,000 cells were seeded into petri dishes 5 cm in diameter. After reaching confluency, the cells were treated with TCHQ or H2 O2 as described above. After treatment of fibroblasts and two washes with ice-cold PBS buffer the cells were trypsinized and resuspended in 0.3 ml ice-cold PBS buffer. An amount of 10 l of the resuspended cells were mixed with 90 l 0.5% low melting agarose (FMC Bio Products, Rockland, USA) at 37 ◦ C and added to the microscope slides prepared with SeaKem agarose LE (FMC Bio Products, Rockland, USA) a day before. Preparing of slides involved ethanol cleaning and dipping into a 1.5% agarose solution diluted in Ca2+ - and Mg2+ -free PBS (pH 7.4). Each concentration was performed in duplicate. The slides mounted with cells were covered with a coverslip and kept in the refrigerator for 3–5 min to solidify the low melting agarose. The following steps were performed under dim light to prevent additional DNA damage. The coverslips were removed and the slides were immersed in refrigerated cold lysing solution, pH 10.0 (2.5 M NaCl, 100 mM EDTA, 10 mM Tris, 1% N-lauroyl sarcosine, 1% Triton X-100, 10% DMSO; the last two components were added freshly). Slides were kept at 4 ◦ C for 1 h. After lysis, the slides were placed on a horizontal electrophoresis box. The
161
unit was filled with freshly prepared alkaline buffer (300 mM NaOH, 1 mM EDTA, pH 13) whereby the slides were completely covered with the buffer. The cells remained at 4 ◦ C for 40 min in the alkaline buffer to allow DNA unwinding and DNA breakage at alkali labile sites. Thereafter, DNA electrophoresis was performed in an ice bath at 25 V and 300 mA for 20 min. After electrophoresis, the slides were horizontally placed and covered with neutralization buffer (0.4 M Tris–HCl, pH 7.5) for 5 min. This step was repeated two times. Thereafter, the slides were shortly dipped into water and dried by air, overnight. Finally, 50 l ethidium bromide (20 g/ml) was added to each slide, covered with a coverslip and kept for 15 min in the dark. DNA migration was analyzed by fluorescence microscopy (Nikon, Fluophot). The tail moment (tm) was determined using the software package “comet assay II” (Perceptive Instruments, Suffolk, UK). The tm considers the length of the tail as well as the intensity of the fluorescence staining of the tail compared to the staining of the comet core. From each concentration, 50 randomly selected cells (25 cells from each of two duplicate slides) were analyzed. The value obtained from untreated controls was about 1 tm. Because comet formation does not follow a Gaussian distribution the significance level by calculation of the standard deviation cannot be determined here. Cells with extremely fragmented DNA visible as a “comet cloud” without a nuclear core cannot be analyzed with the software. They were manually counted in addition to 50 cells with a definable tail moment. The percent ratio of these cells to all scored cells was determined. Determining the persistence of DNA breakage, the comet assay was also applied. After treatment with H2 O2 or TCHQ, the cells were twice washed with serum-free medium. Thereafter, they were allowed to recover in serum containing medium for 0–24 h at 37 ◦ C. After the chosen time intervals, the cells were washed and lysed and the comet assay was performed as described above. 2.5. Determination of DNA repair (unscheduled DNA synthesis) DNA repair was measured by UDS according to Butterworth et al. [15]. In brief, 4 × 8000 cells were seeded on chamber slides with four chambers. After
162
M. Purschke et al. / Mutation Research 513 (2002) 159–167
reaching subconfluency, the cells were incubated with the chemicals as described above. The incubation mix contained 10 Ci [3 H]-thymidine/ml to enable thymidine incorporation. After 1 h treatment with the agents as described above and washing of the cells, an additional incubation with 10 Ci [3 H]-thymidine/ml medium for 3 h at 37 ◦ C was performed followed by a 2 h recovery phase in medium. Fixation of the cells and autoradiography were performed as described by Butterworth et al. [15]. The grains over the nucleus and over an equivalent area of the cytoplasm were computer assisted analyzed using the software Grain V 1.6 (York Electronic Research). For each sample, 50 randomly selected cells were analyzed. The UDS was defined as positive if at least five net (five grains counted over the nucleus than over the cytoplasm) were observed. Positive controls were gained by irradiating cells with UV light (5.5 J/cm2 ). 2.6. Mutagenicity testing (HPRT assay) The HPRT assay was performed according to Bradley et al. [16]. Briefly, V79 cells were seeded in Roux flasks (175 cm2 ). When cells reached subconfluency they were treated with the chemicals for 1 h as described above. Thereafter, 250 trypsinized cells of each sample were seeded in petri dishes for measuring colony forming ability (cfa). In all, 107
cells remained in the flasks. During the following expression period of 7 days, cells were trypsinized and counted at 36–48 h intervals and diluted to a cell number of 107 per flask. For selection of mutants 5 × 105 cells were suspended in medium containing 10 g/ml 6-thioguanine and seeded in petri dishes (150 cm2 ). After 7 days colonies were counted after fixation with formaldehyde and staining with crystal violet. Mutant frequency is expressed as mutants per 106 clonable cells. As a positive control, 10 mM ethylmethane sulfonate (EMS) was used.
3. Results DNA damage in human fibroblasts was determined after a 1 h treatment with H2 O2 or TCHQ by the comet assay. H2 O2 as well as TCHQ induced DNA breakage in a concentration dependent manner (Fig. 1). While H2 O2 induced DNA damage at concentrations >20 M, TCHQ was effective at lower concentrations (≥5 M). The greater DNA damaging potential of TCHQ was also seen in the concentration needed to reach a distinct tail moment. A tail moment of 20 was induced by <10 M TCHQ whereas 60 M H2 O2 produced the same effect. The autoxidation product of TCHQ, tetrachlorobenzoquinone (TCBQ), induced DNA breakage in concentrations of 20 M and up
Fig. 1. DNA damage in human fibroblasts measured by the comet assay. The tail moment was measured after a 1 h treatment of the cells with TCHQ (䉬), H2 O2 (䉫) or TCBQ (䊏).
M. Purschke et al. / Mutation Research 513 (2002) 159–167
163
Fig. 2. Persistence of the DNA damage after a 1 h treatment with TCHQ (䉬) or H2 O2 (䉫). The tail moment was measured after a recovery phase of 15 min until 24 h after treatment with the agents.
(Fig. 1). At 50 M TCBQ provoked the same DNA damage as 7.5 M TCHQ. The cells were allowed to repair the induced DNA damage for up to 24 h after treatment with H2 O2 or TCHQ. As shown in Fig. 2, TCHQ-induced DNA damage was more persistent than H2 O2 -induced damage. Within 30 min after treatment with H2 O2 , the tail moment was reduced by 50% and after 6 h of repair no DNA damage was observed. In contrast, TCHQ-induced damage was not reduced during the first 30 min after the end of treatment, and after 24 h 15% of the initial DNA damage persisted. Measurement of persistence of the DNA damage represents an indirect method for determining DNA repair because only the decrease of DNA damage is determined. The incorporation of [3 H]-thymidine into DNA of non-replicating cells (UDS) measures DNA repair directly. Using this method, it was shown that 2–10 M TCHQ provoked an increase of [3 H]-thymidine incorporation into cellular DNA (Fig. 3). At higher concentrations, the number of net grains decreased. At 25 M, no DNA repair was observed any longer. In contrast, H2 O2 continuously induced DNA repair up to 60 M (Fig. 3). In both
cases [3 H]-thymidine incorporation was low (8–10 net grains). We tested if the observed DNA damage by TCHQ and H2 O2 resulted in mutagenicity. The HPRT assay to determine the mutant frequency in V79 cells was used. In this test, a three-fold increase of mutations over the control is by definition a mutagenic occurrence [16]. The results in Table 1 show that 70 and 110 M H2 O2 are not mutagenic in V79 cells. TCHQ, however, Table 1 Mutagenicity of H2 O2 and TCHQ measured by the HPRT assay in V79 cells Treatment
Untreated control H2 O2 (M) 70 110 TCHQ (M) 5 7 EMS (10 mM)
Mutant frequency (mutants/106 clonable cells)
Colony forming ability (%)
17.9 ± 9.5
100.0 ± 3.0
29.9 ± 7.5 37.9 ± 20.9
143.7 ± 4.5 103.3 ± 1.7
74.6 ± 7.5 151.0 ± 16.9
98.3 ± 13.1 117.9 ± 16.4
243.3 ± 8.5
114.6 ± 7.5
164
M. Purschke et al. / Mutation Research 513 (2002) 159–167
Fig. 3. DNA repair of TCHQ (䉬)- and H2 O2 (䉫)-induced DNA damage measured by UDS. The net grains represent the average number of 50 nuclei counted. The positive UV control (5.5 J/cm2 ) induced 36 ± 5% net grains.
showed mutagenic activity with a mutant frequency of 75 and 151 mutants/106 clonable cells at the non-toxic concentrations of 5 and 7 M (Table 1). Cytotoxicity of TCHQ and H2 O2 was measured with the MTT assay. The cell viability was determined directly and 21 h after incubation with the chemicals. As seen in Fig. 4, H2 O2 was not toxic
in concentrations up to 50 M, neither directly nor 21 h after treatment. In contrast, TCHQ was toxic, whereby the toxicity after a 21 h recovery phase increased. Directly after treatment, about 50 M TCHQ reduced cell viability by 50%, 21 h after treatment the same effect was achieved by half of this concentration.
Fig. 4. Cell viability of human fibroblasts after a 1 h treatment with TCHQ (䉬, 䊏) and H2 O2 (䉫, 䊐). Cell viability directly after treatment, rhombus symbol; cell viability after a recovery phase of 21 h, square symbol.
M. Purschke et al. / Mutation Research 513 (2002) 159–167
4. Discussion The results of treatment of fibroblasts with H2 O2 or TCHQ showed major differences in the kinetics of DNA damage, the repair of the induced DNA damage, the persistence of this damage, and in the mutagenicity as well as the cytotoxicity. One of the differences was demonstrated in the degree of DNA damage. This was shown by the concentrations needed for comet formation (for TCHQ 6 M and up; for H2 O2 20 M and up) and in its frequency (for TCHQ 5 tm/M; for H2 O2 0.5 tm/M). Lower concentrations of TCHQ than of H2 O2 resulted in more damage. The H2 O2 -induced DNA damage was less persistent than the TCHQ-induced DNA damage. Other authors also observed a total removal of H2 O2 -induced comets within 1 h [17,18]. Persistence of DNA damage after TCHQ treatment points to slower repair. UDS, which is used as a direct method to demonstrate DNA excision repair, is only sensitive for long patch repair. In cases where only a few bases are excised, the UDS is not a sensitive method to detect DNA repair. This is true for H2 O2 -induced DNA repair, where the 3 H-incorporation observed was low [19,20]. However, in our experiments, H2 O2 induced in non-toxic concentrations more net grains than TCHQ. TCHQ-induced UDS, decreased at concentrations higher than 10 M. The cytotoxicity may not be the cause for the decrease because a 75% reduction of DNA repair occurred only at a slightly toxic concentration of 15 M TCHQ. The decrease rather points to DNA repair inhibition by the DNA damaging TCHQ molecule itself. This is also in agreement with the prolonged persistence of TCHQ-induced DNA damage compared to H2 O2 . A consequence of unrepaired DNA damage or delayed repair may be the pronounced mutagenicity of TCHQ. The mutagenicity of TCHQ has been demonstrated earlier [21]. Jansson and Jansson observed mutagenicity in toxic concentrations of TCHQ [21], while we observed mutagenic activity at non-toxic concentrations. Mutations were detected at the same concentration range where inhibition of DNA repair occurred. It is, therefore, also possible, that the mutagenicity of TCHQ is a consequence of the observed repair inhibition by TCHQ.
165
H2 O2 did not induce mutants when concentrations five-fold over effective DNA damaging concentrations were used. This is in agreement with other authors [22] and can be explained by the effective repair of H2 O2 -induced DNA damage. H2 O2 can also inhibit DNA repair systems under special circumstances. For example, an inhibitory effect of H2 O2 was observed in the DNA repair of N-acetoxy-2-acetylaminofluorene- [23] and of N-methyl-N -nitro-nitrosoguanidineinduced damage [24]. However, while TCHQ inhibited the repair of its own DNA damage while H2 O2 did not. The delayed cytotoxicity shown 21 h after TCHQ treatment may also be the result of persistent DNA damages, because protein synthesis needed within this time is most likely reduced. But for H2 O2 the recovery phase of 21 h is sufficient to repair all DNA damages (Fig. 2). This explains why there are no differences in cytotoxicity directly after H2 O2 treatment and 21 h later. In previous experiments we had shown that TCHQ causes strand breaks in superhelical PM2 DNA [1]. To cause effects we had to use 4–10-fold higher concentrations than in cell cultures [1]. This may be explained in two ways: 1. In the cell, the hydroquinone may undergo redox cycling to the corresponding quinone and back to the hydroquinone [25,26]. In this cycle, one TCHQ molecule would be able to damage DNA repeatedly. The reduction of the TCBQ cannot take place in the experiments with isolated DNA because the reduction is performed by enzymes or by reducing equivalents (e.g. NADPH) and transition metal ions (e.g. Cu(II)) [4,11]. Neither enzymes nor reducing equivalents were present in our incubation mix with PM2 DNA. If redox cycling would be responsible for lower DNA damaging concentrations of TCHQ in the cell, an incubation with TCBQ at the same concentrations as TCHQ should also result in DNA damage. This was not the case (Fig. 1). Therefore, redox cycling cannot be responsible for enhanced DNA damage in the cell compared to isolated DNA. 2. The observed DNA damage in PM2 DNA is attributed exclusively to directly induced strand breaks, for example by hydroxyl radicals. In the comet assay, additionally alkali labile sites as well as repair induced incisions are detected. Covalent
166
M. Purschke et al. / Mutation Research 513 (2002) 159–167
binding of TCHQ or its corresponding semiquinone to DNA may lead to AP sites which are transformed to strand breaks either by endonucleases or under the alkaline conditions of the comet assay. This would also explain the ineffectiveness of DMSO on TCHQ-induced damage in cellular DNA [7]. If hydroxyl radicals are excluded as ultimate damaging species of TCHQ within the cell, the semiquinone TCSQ becomes the candidate for sustainable DNA damage. Recently, it was shown that TCHQ-induced DNA damage in human fibroblasts was totally inhibited by desferrioxamine [7]. The effect of desferrioxamine was not the result of scavenging iron to abolish • OH radical formation in the Fenton reaction but of scavenging the reactive TCSQ [2,7]. These findings also point to TCSQ as main DNA damaging molecule in TCHQ-induced genotoxicity in human cells. In summary, the observed highly toxic and mutagenic potential of the PCP metabolite TCHQ is not the result of DNA damage by • OH radicals, but it is most likely due to its autoxidation intermediate tetrachlorosemiquinone. Additionally, inhibition of DNA repair by TCHQ allows the DNA damage to survive for a period sufficient for the formation of mutations.
Acknowledgements We thank Dr. Ursula Juhl-Strauss for critical reading of the manuscript.
References [1] C.P. Carstens, J.K. Blum, I. Witte, The role of hydroxyl radicals in tetrachlorohydroquinone induced DNA strand break formation in PM2 DNA and human fibroblasts, Chem. Biol. Interact. 74 (1990) 305–314. [2] B.Z. Zhu, R. Har-El, N. Kitrossky, M. Chevion, New modes of action of desferrioxamine: scavenging of semiquinone radical and stimulation of hydrolysis of tetrachlorohydroquinone, Free Rad. Biol. Med. 24 (1998) 360–369. [3] S. Naito, Y. Ono, I. Somiya, S. Inoue, K. Ito, K. Yamamoto, S. Kawanishi, Role of active oxygen species in DNA damage by pentachlorophenol metabolites, Mutat. Res. 310 (1994) 79–88. [4] M. Dahlhaus, E. Almstadt, P. Henschke, S. Lüttgert, K.E. Appel, Oxidative DNA lesions in V79 cells mediated by pentachlorophenol metabolites, Arch. Toxicol. 70 (1996) 457–460.
[5] M. Dahlhaus, E. Almstadt, P. Henschke, S. Luttgert, K.E. Appel, Induction of 8-hydroxy-2-deoxyguanosine and singlestrand breaks in DNA of V79 cells by tetrachloro-p-hydroquinone, Mutat. Res. 329 (1995) 29–36. [6] M. Dahlhaus, E. Almstadt, K.E. Appel, The pentachlorophenol metabolite tetrachloro-p-hydroquinone induces the formation of 8-hydroxy-2-deoxyguanosine in liver DNA of male B6C3F1 mice, Toxicol. Lett. 74 (1994) 265–274. [7] I. Witte, B.Z. Zhu, A. Lueken, D. Magnani, H. Stossberg, M. Chevion, Protection by desferrioxamine and other hydroxamic acids against tetrachlorohydroquinone-induced cyto- and genotoxicity in human fibroblasts, Free Rad. Biol. Med. 28 (2000) 693–700. [8] J.F. Ward, W.F. Blakely, E.I. Joner, Mammalian cells are not killed by DNA single-strand breaks caused by hydroxyl radicals from hydrogen peroxide, Radiat. Res. 103 (1985) 383–392. [9] I. Witte, U. Juhl, W. Butte, DNA-damaging properties and cytotoxicity in human fibroblasts of tetrachlorohydroquinone: a pentachlorophenol metabolite, Mutat. Res. 145 (1985) 71–75. [10] B. Van Ommen, A. Adang, F. Muller, P.J. Van Bladeren, The microsomal metabolism of pentachlorophenol and its covalent binding to protein and DNA, Chem. Biol. Interact. 60 (1986) 1–11. [11] P.-H. Lin, J. Nakamura, S. Yamaguchi, P.B. Upton, D.K. La, J.A. Swenberg, Oxidative damage and direct adducts in calf thymus DNA induced by pentachlorophenol metabolites, tetrachlorohydroquinone and tetrachloro-1,4-benzoquinone, Carcinogenesis 22 (2001) 627–634. [12] P.-H. Lin, J. Nakamura, S. Yamaguchi, D.K. La, P.B. Upton, J.A. Swenberg, Induction of direct adducts, apurinic/ apyrimidinic sites and oxidized bases in nuclear DNA of human HeLa S3 tumor cells by tetrachlorohydroquinone, Carcinogenesis 22 (2001) 635–639. [13] M.B. Hansen, S.E. Nielsen, K. Berg, Re-examination and further development of a precise and rapid dye method for measuring cell growth/cell kill, J. Immunol. Methods 119 (1989) 203–210. [14] N.P. Singh, M.T. McCoy, R.R. Tice, E.L. Schneider, A simple technique for quantitation of low levels of DNA damage in individual cells, Exp. Cell Res. 175 (1988) 184–191. [15] B.E. Butterworth, J. Ashby, E. Bermudez, D. Casciano, J. Mirsalis, G. Probst, G. Williams, A protocol and guide for the in vitro rat hepatocyte DNA-repair assay, Mutat. Res. 189 (1987) 113–121. [16] M.O. Bradley, B. Bhuyan, M.C. Francis, R. Langenbach, A. Peterson, E. Huberman, Mutagenesis by chemical agents in V79 Chinese hamster cells: a review and analysis of the literature, Mutat. Res. 87 (1981) 81–142. [17] A.R. Collins, A.G. Ma, S.J. Duthie, The kinetics of repair of oxidative DNA damage (strand breaks and oxidised pyrimidines) in human cells, Mutat. Res. 336 (1) (1995) 69–77. [18] M. Kruszewski, M.H.L. Green, J.E. Lowe, I. Szumiel, DNA strand breakage, cytotoxicity and mutagenicity of hydrogen peroxide treatment at 4 and 37 ◦ C in L5178Y sublines, Mutat. Res. 308 (1994) 233–241.
M. Purschke et al. / Mutation Research 513 (2002) 159–167 [19] H.F. Stich, L. Wie, P. Lam, The need for a mammalian test system for mutagens: action of some reducing agents, Cancer Lett. 5 (1978) 199–204. [20] W.J. Coppinger, T.K. Wong, E.D. Thompson, Unscheduled DNA synthesis and DNA repair studies of peroxyacetic and monoperoxydecanoic acids, Environ. Mutagen. 5 (1983) 177–192. [21] K. Jansson, V. Jansson, Induction of mutation in V79 Chinese hamster cells by tetrachlorohydroquinone, a metabolite of pentachlorophenol, Mutat. Res. 260 (1991) 83–87. [22] IARC, Hydrogen peroxide, Monogr. Eval. Carcinogen. Risks Hum. 71 (1999) 671–689. [23] R.W. Pero, M.W. Anderson, G.A. Doyle, C.H. Anna, F. Romagna, M. Markowitz, C. Bryngelsson, Oxidative stress
167
induces DNA damage and inhibits the repair of DNA lesions induced by N-acetoxy-2-acetylaminofluorene in human peripheral mononuclear leukocytes, Cancer Res. 50 (15) (1990) 4619–4625. [24] J.J. Hu, N. Dubin, D. Kurland, B.L. Ma, G.C. Roush, The effects of hydrogen peroxide on DNA repair activities, Mutat. Res. 336 (1995) 193–201. [25] A. Brunmark, E. Cadenas, Redox and addition chemistry of quinoid compounds and its biological implications, Free Rad. Biol. Med. 7 (1989) 435–477. [26] P.L. Gutierrez, The metabolism of quinone-containing alkylating agents: free radical production and measurement, Front Biosci. 5 (2000) 629–638.
Differences in genotoxicity of H2 O2 and tetrachlorohydroquinone in human fibroblasts Martin Purschke, Heike Jacobi, Irene Witte∗ Carl von Ossietzky Universität Oldenburg, FB Biologie and ICBM, Postfach 2503, D-26111 Oldenburg, Germany Received 19 June 2001; received in revised form 30 August 2001; accepted 7 September 2001
Abstract During autoxidation of the pentachlorophenol (PCP) metabolite tetrachlorohydroquinone (TCHQ) the semiquinone is formed as well as reactive oxygen species (ROS). It was examined if • OH or the semiquinone are the cause of TCHQ-induced genotoxicity by direct comparison of TCHQ- and H2 O2 -induced DNA damage in human cells. All endpoints tested (DNA damage, DNA repair, and mutagenicity) revealed a greater genotoxic potential for TCHQ than for H2 O2 . In the comet assay, TCHQ induced DNA damage at lower concentrations than H2 O2 . The damaging rate by TCHQ (tail moment (tm)/concentration) was 10-fold greater than by H2 O2 . DNA repair was lower for TCHQ than for H2 O2 treatment. This was shown by measuring DNA repair in the unscheduled DNA synthesis (UDS) assay and the persistence of the DNA damage in the comet assay. In contrast to H2 O2 , TCHQ in non-toxic concentrations was mutagenic in the hypoxanthine-guanine phosphoribosyltransferase (HPRT) locus of V79 cells. Finally, there were also differences observed in cytotoxicity (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay) of TCHQ and H2 O2 . Whereas the TCHQ cytotoxicity was enhanced during a 21 h recovery phase, the H2 O2 cytotoxicity did not change. The results demonstrated that the pronounced genotoxic properties of TCHQ in human cells were not caused by • OH radicals but more likely by the tetrachlorosemiquinone (TCSQ) radical. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Hydrogen peroxide; Tetrachlorohydroquinone; Mutagenicity; Genotoxicity; DNA repair; Cytotoxicity
1. Introduction Tetrachlorohydroquinone (TCHQ) has been identified as the main DNA damaging metabolite of Abbreviations: cfa, colony forming ability; DMSO, dimethyl sulfoxide; EMS, ethylmethane sulfonate; HPRT, hypoxanthine-guanine phosphoribosyltransferase; MEM, minimal essential medium; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide; PBS, phosphate-buffered saline; PCP, pentachlorophenol; ROS, reactive oxygen species; TCBQ, tetrachlorobenzoquinone; TCHQ, tetrachlorohydroquinone; TCSQ, tetrachlorosemiquinone radical; tm, tail moment; UDS, unscheduled DNA synthesis ∗ Corresponding author. Tel.: +49-441-798-3785; fax: +49-441-789-4903. E-mail address: [email protected] (I. Witte).
the widely used and ubiquitously found wood preservative pentachlorophenol (PCP). TCHQ autoxidizes in two one-electron steps to its tetrachlorosemiquinone (TCSQ) radical and further to its quinone [1,2]. The resulting free electrons reduce oxygen to reactive oxygen species (ROS). Genotoxic events of TCHQ may either be due to ROS or TCSQ. While there is little knowledge on TCSQ-induced DNA damage, ROSinduced DNA damage as a consequence of TCHQ autoxidation has been observed repeatedly. Of the ROS, especially the • OH radical is suspected to be the main damaging species in TCHQ genotoxicity. The formation of the typical • OH radical adduct 8hydroxy-2-deoxyguanosine was detected in vitro [3], in cell culture [4,5] and in mice [6] after incubation
1383-5718/02/$ – see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S 1 3 8 3 - 5 7 1 8 ( 0 1 ) 0 0 3 1 1 - 4
160
M. Purschke et al. / Mutation Research 513 (2002) 159–167
or administration of TCHQ. The complete inhibition of TCHQ-induced strand breaks in PM2 DNA by catalase and the specific scavenger dimethyl sulfoxide (DMSO) [1] pointed to • OH radicals as the only cause of DNA strand breakage in superhelical DNA. Nevertheless, there exists some doubts on the dominant role of • OH radicals in TCHQ-induced genotoxicity. In human fibroblasts, the TCHQ-induced DNA damage measured by the alkaline elution [1] or by the comet assay [7] was not completely inhibited by DMSO, even if high concentrations of DMSO were used. In contrast to TCHQ, • OH radicals generated by H2 O2 induced strand breaks which were totally prevented by DMSO in fibroblasts [8]. Further, for TCHQ or its autoxidation products, the covalent binding to DNA was previously shown [9–12]. The genotoxic consequences of those DNA adducts are unknown so far. From these findings, it should be expected that besides ROS also TCHQ itself or its autoxidation product TCSQ are candidates for inducing genotoxic effects in human cells. The aim of this study was to elucidate in mammalian cells the role of DNA damaging and mutagenic species others than ROS in TCHQ genotoxicity. Therefore, a comparison was made of H2 O2 as an exclusively hydroxyl radical generating molecule and TCHQ which additionally forms the reactive semiquinone and quinone molecule. It is expected that DNA damages and DNA repair induced by TCSQ are different from that of • OH radicals. Especially, DNA repair of TCHQ–DNA adducts should follow other pathways than DNA repair of single strand breaks induced by hydroxyl radicals. DNA damage was measured in human fibroblasts by the comet assay. DNA repair was examined with two methods: the persistence of DNA damage with the comet assay, and the 3 [H]-thymidine incorporation into damaged DNA by the method of the unscheduled DNA synthesis (UDS). The consequences of DNA damage and DNA repair were determined by measuring mutagenicity of H2 O2 and TCHQ in the hypoxanthine-guanine phosphoribosyltransferase (HPRT) assay. In addition, the role of DNA damage in cytotoxicity of H2 O2 and TCHQ was examined by comparing the acute toxicity directly after treatment with the chemicals and after a recovery phase of 21 h. While DNA damage rarely should effect cytotoxicity directly after chemical treatment, a reduced cell
viability is expected hours later if unrepaired DNA damage inhibits the protein biosynthesis. 2. Materials and methods 2.1. Cell cultures Human fibroblasts from apparently normal persons, cell lines GM 5757, GM 5856, GM 5659 were purchased from the Human Genetic Mutant Cell Repository (Camden, NJ). Monolayer cultures (passage 7–15) were grown in Eagle’s minimal essential medium (MEM) supplemented with 12% fetal calf serum, vitamins, essential and non-essential amino acids, with 100 U/ml of both penicillin and streptomycin. The cells were grown at 37 ◦ C in an atmosphere of 5% CO2 and 95% air with more than 95% humidity. 2.2. Chemical treatment of cells Tetrachlorohydroquinone (1 mM; from Sigma, FRG) was freshly dissolved in 100 mM NaPO4 buffer, pH 7.4 and sonicated by ultrasound for 4 min. The TCHQ solution was diluted by serum-free medium to the desired concentrations and immediately after washing the cells were treated with the TCHQ solutions for 1 h at 37 ◦ C. H2 O2 was diluted from a 37% aqueous solution (Acros Organics, NJ) to 5 mM in 100 mM NaPO4 buffer, pH 7.4 immediately before cell treatment. Further dilution occurred in serum-free medium. H2 O2 treatment like TCHQ treatment was performed for 1 h at 37 ◦ C. Thereafter, test solutions were removed and the cells were washed repeatedly. 2.3. Determination of cell viability (MTT assay) Cell viability was measured by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay which was performed according to [13]. The test is based on the reduction of the soluble yellow MTT tetrazolium salt to a blue soluble formazan produced by mitochondrial succinate dehydrogenase. In all, 8000 cells were seeded in each well of a 96-well tissue culture microtiter plate. After cells had achieved confluency, the medium was removed and the cells were treated with the test solutions as described before. At least four wells were used for
M. Purschke et al. / Mutation Research 513 (2002) 159–167
every concentration tested. After 1 h of incubation with TCHQ or H2 O2 , the test solutions were removed and the cells were washed twice. MTT containing medium (100 l), which was freshly prepared by a 1:5 dilution of an MTT stock solution (5 mg/ml phosphate-buffered saline (PBS) buffer) was added to each well. The plate was incubated for an additional 3 h. An amount of 100 l of lysis solution (20 g/l SDS, 50%, v/v DMF, pH adjusted to 4.7 by adding 2.5% (v/v) of a 80% acetic acid and 2.5% 1N HCl) were added to each well and the plate was shaken overnight in the dark to extract and solubilize the formazan. Formation of formazan was measured with a Biorad microplate reader using a 570 nm test wavelength and a 655 nm reference wavelength. Cell viability was calculated as the percent ratio of the absorbance of the samples to the referring control. 2.4. Determination of DNA strand breakage (comet assay) The comet assay was performed according to Singh et al. [14]. In brief, 30,000 cells were seeded into petri dishes 5 cm in diameter. After reaching confluency, the cells were treated with TCHQ or H2 O2 as described above. After treatment of fibroblasts and two washes with ice-cold PBS buffer the cells were trypsinized and resuspended in 0.3 ml ice-cold PBS buffer. An amount of 10 l of the resuspended cells were mixed with 90 l 0.5% low melting agarose (FMC Bio Products, Rockland, USA) at 37 ◦ C and added to the microscope slides prepared with SeaKem agarose LE (FMC Bio Products, Rockland, USA) a day before. Preparing of slides involved ethanol cleaning and dipping into a 1.5% agarose solution diluted in Ca2+ - and Mg2+ -free PBS (pH 7.4). Each concentration was performed in duplicate. The slides mounted with cells were covered with a coverslip and kept in the refrigerator for 3–5 min to solidify the low melting agarose. The following steps were performed under dim light to prevent additional DNA damage. The coverslips were removed and the slides were immersed in refrigerated cold lysing solution, pH 10.0 (2.5 M NaCl, 100 mM EDTA, 10 mM Tris, 1% N-lauroyl sarcosine, 1% Triton X-100, 10% DMSO; the last two components were added freshly). Slides were kept at 4 ◦ C for 1 h. After lysis, the slides were placed on a horizontal electrophoresis box. The
161
unit was filled with freshly prepared alkaline buffer (300 mM NaOH, 1 mM EDTA, pH 13) whereby the slides were completely covered with the buffer. The cells remained at 4 ◦ C for 40 min in the alkaline buffer to allow DNA unwinding and DNA breakage at alkali labile sites. Thereafter, DNA electrophoresis was performed in an ice bath at 25 V and 300 mA for 20 min. After electrophoresis, the slides were horizontally placed and covered with neutralization buffer (0.4 M Tris–HCl, pH 7.5) for 5 min. This step was repeated two times. Thereafter, the slides were shortly dipped into water and dried by air, overnight. Finally, 50 l ethidium bromide (20 g/ml) was added to each slide, covered with a coverslip and kept for 15 min in the dark. DNA migration was analyzed by fluorescence microscopy (Nikon, Fluophot). The tail moment (tm) was determined using the software package “comet assay II” (Perceptive Instruments, Suffolk, UK). The tm considers the length of the tail as well as the intensity of the fluorescence staining of the tail compared to the staining of the comet core. From each concentration, 50 randomly selected cells (25 cells from each of two duplicate slides) were analyzed. The value obtained from untreated controls was about 1 tm. Because comet formation does not follow a Gaussian distribution the significance level by calculation of the standard deviation cannot be determined here. Cells with extremely fragmented DNA visible as a “comet cloud” without a nuclear core cannot be analyzed with the software. They were manually counted in addition to 50 cells with a definable tail moment. The percent ratio of these cells to all scored cells was determined. Determining the persistence of DNA breakage, the comet assay was also applied. After treatment with H2 O2 or TCHQ, the cells were twice washed with serum-free medium. Thereafter, they were allowed to recover in serum containing medium for 0–24 h at 37 ◦ C. After the chosen time intervals, the cells were washed and lysed and the comet assay was performed as described above. 2.5. Determination of DNA repair (unscheduled DNA synthesis) DNA repair was measured by UDS according to Butterworth et al. [15]. In brief, 4 × 8000 cells were seeded on chamber slides with four chambers. After
162
M. Purschke et al. / Mutation Research 513 (2002) 159–167
reaching subconfluency, the cells were incubated with the chemicals as described above. The incubation mix contained 10 Ci [3 H]-thymidine/ml to enable thymidine incorporation. After 1 h treatment with the agents as described above and washing of the cells, an additional incubation with 10 Ci [3 H]-thymidine/ml medium for 3 h at 37 ◦ C was performed followed by a 2 h recovery phase in medium. Fixation of the cells and autoradiography were performed as described by Butterworth et al. [15]. The grains over the nucleus and over an equivalent area of the cytoplasm were computer assisted analyzed using the software Grain V 1.6 (York Electronic Research). For each sample, 50 randomly selected cells were analyzed. The UDS was defined as positive if at least five net (five grains counted over the nucleus than over the cytoplasm) were observed. Positive controls were gained by irradiating cells with UV light (5.5 J/cm2 ). 2.6. Mutagenicity testing (HPRT assay) The HPRT assay was performed according to Bradley et al. [16]. Briefly, V79 cells were seeded in Roux flasks (175 cm2 ). When cells reached subconfluency they were treated with the chemicals for 1 h as described above. Thereafter, 250 trypsinized cells of each sample were seeded in petri dishes for measuring colony forming ability (cfa). In all, 107
cells remained in the flasks. During the following expression period of 7 days, cells were trypsinized and counted at 36–48 h intervals and diluted to a cell number of 107 per flask. For selection of mutants 5 × 105 cells were suspended in medium containing 10 g/ml 6-thioguanine and seeded in petri dishes (150 cm2 ). After 7 days colonies were counted after fixation with formaldehyde and staining with crystal violet. Mutant frequency is expressed as mutants per 106 clonable cells. As a positive control, 10 mM ethylmethane sulfonate (EMS) was used.
3. Results DNA damage in human fibroblasts was determined after a 1 h treatment with H2 O2 or TCHQ by the comet assay. H2 O2 as well as TCHQ induced DNA breakage in a concentration dependent manner (Fig. 1). While H2 O2 induced DNA damage at concentrations >20 M, TCHQ was effective at lower concentrations (≥5 M). The greater DNA damaging potential of TCHQ was also seen in the concentration needed to reach a distinct tail moment. A tail moment of 20 was induced by <10 M TCHQ whereas 60 M H2 O2 produced the same effect. The autoxidation product of TCHQ, tetrachlorobenzoquinone (TCBQ), induced DNA breakage in concentrations of 20 M and up
Fig. 1. DNA damage in human fibroblasts measured by the comet assay. The tail moment was measured after a 1 h treatment of the cells with TCHQ (䉬), H2 O2 (䉫) or TCBQ (䊏).
M. Purschke et al. / Mutation Research 513 (2002) 159–167
163
Fig. 2. Persistence of the DNA damage after a 1 h treatment with TCHQ (䉬) or H2 O2 (䉫). The tail moment was measured after a recovery phase of 15 min until 24 h after treatment with the agents.
(Fig. 1). At 50 M TCBQ provoked the same DNA damage as 7.5 M TCHQ. The cells were allowed to repair the induced DNA damage for up to 24 h after treatment with H2 O2 or TCHQ. As shown in Fig. 2, TCHQ-induced DNA damage was more persistent than H2 O2 -induced damage. Within 30 min after treatment with H2 O2 , the tail moment was reduced by 50% and after 6 h of repair no DNA damage was observed. In contrast, TCHQ-induced damage was not reduced during the first 30 min after the end of treatment, and after 24 h 15% of the initial DNA damage persisted. Measurement of persistence of the DNA damage represents an indirect method for determining DNA repair because only the decrease of DNA damage is determined. The incorporation of [3 H]-thymidine into DNA of non-replicating cells (UDS) measures DNA repair directly. Using this method, it was shown that 2–10 M TCHQ provoked an increase of [3 H]-thymidine incorporation into cellular DNA (Fig. 3). At higher concentrations, the number of net grains decreased. At 25 M, no DNA repair was observed any longer. In contrast, H2 O2 continuously induced DNA repair up to 60 M (Fig. 3). In both
cases [3 H]-thymidine incorporation was low (8–10 net grains). We tested if the observed DNA damage by TCHQ and H2 O2 resulted in mutagenicity. The HPRT assay to determine the mutant frequency in V79 cells was used. In this test, a three-fold increase of mutations over the control is by definition a mutagenic occurrence [16]. The results in Table 1 show that 70 and 110 M H2 O2 are not mutagenic in V79 cells. TCHQ, however, Table 1 Mutagenicity of H2 O2 and TCHQ measured by the HPRT assay in V79 cells Treatment
Untreated control H2 O2 (M) 70 110 TCHQ (M) 5 7 EMS (10 mM)
Mutant frequency (mutants/106 clonable cells)
Colony forming ability (%)
17.9 ± 9.5
100.0 ± 3.0
29.9 ± 7.5 37.9 ± 20.9
143.7 ± 4.5 103.3 ± 1.7
74.6 ± 7.5 151.0 ± 16.9
98.3 ± 13.1 117.9 ± 16.4
243.3 ± 8.5
114.6 ± 7.5
164
M. Purschke et al. / Mutation Research 513 (2002) 159–167
Fig. 3. DNA repair of TCHQ (䉬)- and H2 O2 (䉫)-induced DNA damage measured by UDS. The net grains represent the average number of 50 nuclei counted. The positive UV control (5.5 J/cm2 ) induced 36 ± 5% net grains.
showed mutagenic activity with a mutant frequency of 75 and 151 mutants/106 clonable cells at the non-toxic concentrations of 5 and 7 M (Table 1). Cytotoxicity of TCHQ and H2 O2 was measured with the MTT assay. The cell viability was determined directly and 21 h after incubation with the chemicals. As seen in Fig. 4, H2 O2 was not toxic
in concentrations up to 50 M, neither directly nor 21 h after treatment. In contrast, TCHQ was toxic, whereby the toxicity after a 21 h recovery phase increased. Directly after treatment, about 50 M TCHQ reduced cell viability by 50%, 21 h after treatment the same effect was achieved by half of this concentration.
Fig. 4. Cell viability of human fibroblasts after a 1 h treatment with TCHQ (䉬, 䊏) and H2 O2 (䉫, 䊐). Cell viability directly after treatment, rhombus symbol; cell viability after a recovery phase of 21 h, square symbol.
M. Purschke et al. / Mutation Research 513 (2002) 159–167
4. Discussion The results of treatment of fibroblasts with H2 O2 or TCHQ showed major differences in the kinetics of DNA damage, the repair of the induced DNA damage, the persistence of this damage, and in the mutagenicity as well as the cytotoxicity. One of the differences was demonstrated in the degree of DNA damage. This was shown by the concentrations needed for comet formation (for TCHQ 6 M and up; for H2 O2 20 M and up) and in its frequency (for TCHQ 5 tm/M; for H2 O2 0.5 tm/M). Lower concentrations of TCHQ than of H2 O2 resulted in more damage. The H2 O2 -induced DNA damage was less persistent than the TCHQ-induced DNA damage. Other authors also observed a total removal of H2 O2 -induced comets within 1 h [17,18]. Persistence of DNA damage after TCHQ treatment points to slower repair. UDS, which is used as a direct method to demonstrate DNA excision repair, is only sensitive for long patch repair. In cases where only a few bases are excised, the UDS is not a sensitive method to detect DNA repair. This is true for H2 O2 -induced DNA repair, where the 3 H-incorporation observed was low [19,20]. However, in our experiments, H2 O2 induced in non-toxic concentrations more net grains than TCHQ. TCHQ-induced UDS, decreased at concentrations higher than 10 M. The cytotoxicity may not be the cause for the decrease because a 75% reduction of DNA repair occurred only at a slightly toxic concentration of 15 M TCHQ. The decrease rather points to DNA repair inhibition by the DNA damaging TCHQ molecule itself. This is also in agreement with the prolonged persistence of TCHQ-induced DNA damage compared to H2 O2 . A consequence of unrepaired DNA damage or delayed repair may be the pronounced mutagenicity of TCHQ. The mutagenicity of TCHQ has been demonstrated earlier [21]. Jansson and Jansson observed mutagenicity in toxic concentrations of TCHQ [21], while we observed mutagenic activity at non-toxic concentrations. Mutations were detected at the same concentration range where inhibition of DNA repair occurred. It is, therefore, also possible, that the mutagenicity of TCHQ is a consequence of the observed repair inhibition by TCHQ.
165
H2 O2 did not induce mutants when concentrations five-fold over effective DNA damaging concentrations were used. This is in agreement with other authors [22] and can be explained by the effective repair of H2 O2 -induced DNA damage. H2 O2 can also inhibit DNA repair systems under special circumstances. For example, an inhibitory effect of H2 O2 was observed in the DNA repair of N-acetoxy-2-acetylaminofluorene- [23] and of N-methyl-N -nitro-nitrosoguanidineinduced damage [24]. However, while TCHQ inhibited the repair of its own DNA damage while H2 O2 did not. The delayed cytotoxicity shown 21 h after TCHQ treatment may also be the result of persistent DNA damages, because protein synthesis needed within this time is most likely reduced. But for H2 O2 the recovery phase of 21 h is sufficient to repair all DNA damages (Fig. 2). This explains why there are no differences in cytotoxicity directly after H2 O2 treatment and 21 h later. In previous experiments we had shown that TCHQ causes strand breaks in superhelical PM2 DNA [1]. To cause effects we had to use 4–10-fold higher concentrations than in cell cultures [1]. This may be explained in two ways: 1. In the cell, the hydroquinone may undergo redox cycling to the corresponding quinone and back to the hydroquinone [25,26]. In this cycle, one TCHQ molecule would be able to damage DNA repeatedly. The reduction of the TCBQ cannot take place in the experiments with isolated DNA because the reduction is performed by enzymes or by reducing equivalents (e.g. NADPH) and transition metal ions (e.g. Cu(II)) [4,11]. Neither enzymes nor reducing equivalents were present in our incubation mix with PM2 DNA. If redox cycling would be responsible for lower DNA damaging concentrations of TCHQ in the cell, an incubation with TCBQ at the same concentrations as TCHQ should also result in DNA damage. This was not the case (Fig. 1). Therefore, redox cycling cannot be responsible for enhanced DNA damage in the cell compared to isolated DNA. 2. The observed DNA damage in PM2 DNA is attributed exclusively to directly induced strand breaks, for example by hydroxyl radicals. In the comet assay, additionally alkali labile sites as well as repair induced incisions are detected. Covalent
166
M. Purschke et al. / Mutation Research 513 (2002) 159–167
binding of TCHQ or its corresponding semiquinone to DNA may lead to AP sites which are transformed to strand breaks either by endonucleases or under the alkaline conditions of the comet assay. This would also explain the ineffectiveness of DMSO on TCHQ-induced damage in cellular DNA [7]. If hydroxyl radicals are excluded as ultimate damaging species of TCHQ within the cell, the semiquinone TCSQ becomes the candidate for sustainable DNA damage. Recently, it was shown that TCHQ-induced DNA damage in human fibroblasts was totally inhibited by desferrioxamine [7]. The effect of desferrioxamine was not the result of scavenging iron to abolish • OH radical formation in the Fenton reaction but of scavenging the reactive TCSQ [2,7]. These findings also point to TCSQ as main DNA damaging molecule in TCHQ-induced genotoxicity in human cells. In summary, the observed highly toxic and mutagenic potential of the PCP metabolite TCHQ is not the result of DNA damage by • OH radicals, but it is most likely due to its autoxidation intermediate tetrachlorosemiquinone. Additionally, inhibition of DNA repair by TCHQ allows the DNA damage to survive for a period sufficient for the formation of mutations.
Acknowledgements We thank Dr. Ursula Juhl-Strauss for critical reading of the manuscript.
References [1] C.P. Carstens, J.K. Blum, I. Witte, The role of hydroxyl radicals in tetrachlorohydroquinone induced DNA strand break formation in PM2 DNA and human fibroblasts, Chem. Biol. Interact. 74 (1990) 305–314. [2] B.Z. Zhu, R. Har-El, N. Kitrossky, M. Chevion, New modes of action of desferrioxamine: scavenging of semiquinone radical and stimulation of hydrolysis of tetrachlorohydroquinone, Free Rad. Biol. Med. 24 (1998) 360–369. [3] S. Naito, Y. Ono, I. Somiya, S. Inoue, K. Ito, K. Yamamoto, S. Kawanishi, Role of active oxygen species in DNA damage by pentachlorophenol metabolites, Mutat. Res. 310 (1994) 79–88. [4] M. Dahlhaus, E. Almstadt, P. Henschke, S. Lüttgert, K.E. Appel, Oxidative DNA lesions in V79 cells mediated by pentachlorophenol metabolites, Arch. Toxicol. 70 (1996) 457–460.
[5] M. Dahlhaus, E. Almstadt, P. Henschke, S. Luttgert, K.E. Appel, Induction of 8-hydroxy-2-deoxyguanosine and singlestrand breaks in DNA of V79 cells by tetrachloro-p-hydroquinone, Mutat. Res. 329 (1995) 29–36. [6] M. Dahlhaus, E. Almstadt, K.E. Appel, The pentachlorophenol metabolite tetrachloro-p-hydroquinone induces the formation of 8-hydroxy-2-deoxyguanosine in liver DNA of male B6C3F1 mice, Toxicol. Lett. 74 (1994) 265–274. [7] I. Witte, B.Z. Zhu, A. Lueken, D. Magnani, H. Stossberg, M. Chevion, Protection by desferrioxamine and other hydroxamic acids against tetrachlorohydroquinone-induced cyto- and genotoxicity in human fibroblasts, Free Rad. Biol. Med. 28 (2000) 693–700. [8] J.F. Ward, W.F. Blakely, E.I. Joner, Mammalian cells are not killed by DNA single-strand breaks caused by hydroxyl radicals from hydrogen peroxide, Radiat. Res. 103 (1985) 383–392. [9] I. Witte, U. Juhl, W. Butte, DNA-damaging properties and cytotoxicity in human fibroblasts of tetrachlorohydroquinone: a pentachlorophenol metabolite, Mutat. Res. 145 (1985) 71–75. [10] B. Van Ommen, A. Adang, F. Muller, P.J. Van Bladeren, The microsomal metabolism of pentachlorophenol and its covalent binding to protein and DNA, Chem. Biol. Interact. 60 (1986) 1–11. [11] P.-H. Lin, J. Nakamura, S. Yamaguchi, P.B. Upton, D.K. La, J.A. Swenberg, Oxidative damage and direct adducts in calf thymus DNA induced by pentachlorophenol metabolites, tetrachlorohydroquinone and tetrachloro-1,4-benzoquinone, Carcinogenesis 22 (2001) 627–634. [12] P.-H. Lin, J. Nakamura, S. Yamaguchi, D.K. La, P.B. Upton, J.A. Swenberg, Induction of direct adducts, apurinic/ apyrimidinic sites and oxidized bases in nuclear DNA of human HeLa S3 tumor cells by tetrachlorohydroquinone, Carcinogenesis 22 (2001) 635–639. [13] M.B. Hansen, S.E. Nielsen, K. Berg, Re-examination and further development of a precise and rapid dye method for measuring cell growth/cell kill, J. Immunol. Methods 119 (1989) 203–210. [14] N.P. Singh, M.T. McCoy, R.R. Tice, E.L. Schneider, A simple technique for quantitation of low levels of DNA damage in individual cells, Exp. Cell Res. 175 (1988) 184–191. [15] B.E. Butterworth, J. Ashby, E. Bermudez, D. Casciano, J. Mirsalis, G. Probst, G. Williams, A protocol and guide for the in vitro rat hepatocyte DNA-repair assay, Mutat. Res. 189 (1987) 113–121. [16] M.O. Bradley, B. Bhuyan, M.C. Francis, R. Langenbach, A. Peterson, E. Huberman, Mutagenesis by chemical agents in V79 Chinese hamster cells: a review and analysis of the literature, Mutat. Res. 87 (1981) 81–142. [17] A.R. Collins, A.G. Ma, S.J. Duthie, The kinetics of repair of oxidative DNA damage (strand breaks and oxidised pyrimidines) in human cells, Mutat. Res. 336 (1) (1995) 69–77. [18] M. Kruszewski, M.H.L. Green, J.E. Lowe, I. Szumiel, DNA strand breakage, cytotoxicity and mutagenicity of hydrogen peroxide treatment at 4 and 37 ◦ C in L5178Y sublines, Mutat. Res. 308 (1994) 233–241.
M. Purschke et al. / Mutation Research 513 (2002) 159–167 [19] H.F. Stich, L. Wie, P. Lam, The need for a mammalian test system for mutagens: action of some reducing agents, Cancer Lett. 5 (1978) 199–204. [20] W.J. Coppinger, T.K. Wong, E.D. Thompson, Unscheduled DNA synthesis and DNA repair studies of peroxyacetic and monoperoxydecanoic acids, Environ. Mutagen. 5 (1983) 177–192. [21] K. Jansson, V. Jansson, Induction of mutation in V79 Chinese hamster cells by tetrachlorohydroquinone, a metabolite of pentachlorophenol, Mutat. Res. 260 (1991) 83–87. [22] IARC, Hydrogen peroxide, Monogr. Eval. Carcinogen. Risks Hum. 71 (1999) 671–689. [23] R.W. Pero, M.W. Anderson, G.A. Doyle, C.H. Anna, F. Romagna, M. Markowitz, C. Bryngelsson, Oxidative stress
167
induces DNA damage and inhibits the repair of DNA lesions induced by N-acetoxy-2-acetylaminofluorene in human peripheral mononuclear leukocytes, Cancer Res. 50 (15) (1990) 4619–4625. [24] J.J. Hu, N. Dubin, D. Kurland, B.L. Ma, G.C. Roush, The effects of hydrogen peroxide on DNA repair activities, Mutat. Res. 336 (1995) 193–201. [25] A. Brunmark, E. Cadenas, Redox and addition chemistry of quinoid compounds and its biological implications, Free Rad. Biol. Med. 7 (1989) 435–477. [26] P.L. Gutierrez, The metabolism of quinone-containing alkylating agents: free radical production and measurement, Front Biosci. 5 (2000) 629–638.